|
|
|
26 May 2009
I've revised my
Off-the-air frequency
measurement page to show a simple modification to the HP3586B selective
voltmeter to extract a 15625 Hz 2nd IF output signal. The 15625 Hz signal is
fully locked to the 3586B's time base, or, in my case, to a Trimble Thunderbolt
GPS locked 10 MHz source.
|
|
22 May 2009
I visited Mike, W4XN, in Charlottesville, VA, this week
and a couple of collector's items of vintage test equipment followed me home.
One is a Tektronix 575 curve tracer, of circa 1960 manufacture.
I have a Telequipment CT-71 curve tracer, but it's a
budget-build design and while useful, it drifts more than one might desire and
is difficult to take decent quality photos of the CRT. Although probably 15
years older than the CT-71, in my brief exposure to the 575, it seems
considerably more stable than the CT-71.
|
|
Tektronix 575 showing Ic versus Ib of 2N4403 Silicon PNP
transistor. The Ic versus Ib display shows how DC current gain (β) varies
with collector current.
|
 |
The image below is a not-so-great quality photo of the trace seen in the small
image above. I've annotated it to show what is being measured and the results.
I hand-held the digital camera for this image and the
automatic shutter speed selector picked a speed a bit too slow for best image
quality when hand held. I also have to work on getting the camera centered, as
there is clear parallax error. When seen in person, the vertical base current
lines line up almost perfectly with the graticule. Later Tektronix and HP analog
oscilloscopes have the graticule etched into the inside of the CRT face and
hence there's no parallax error. The 575, like other oscilloscopes of the era,
uses a separate acrylic graticule, so the electron beam image is behind the
graticule by the thickness of the CRT glass face.
The plot shows collector current on the vertical axis and
base current on the horizontal axis. (Since this is a PNP transistor, both the
collector and base current is negative and hence plots in the third quadrant.)
The real information on this plot is the continuous line connecting the ends of
the vertical straight lines, not the vertical straight lines themselves.
The DC current gain, β or HFE, of a
transistor is defined as:
β = ΔIc / ΔIb
ΔIc is the change in collector current caused by a change in base current
ΔIb.
ΔIb in the 575 is selected by the base current step generator. In this case,
I've used 100 μA (0.1 mA) steps. Each step corresponds to one horizontal
graticule line.
Collector current is displayed on the vertical scale, at 20 mA/division. The
collector voltage is 20V, with 100 ohms series resistance.
The choice of ΔIb is therefore 0.1 mA. I've annotated the image with three
tangent lines at (approximately) mid point Ic values of 40, 140 and 180 mA. ΔIc
is the change in collector current corresponding to one base current step. At 40
mA current, the base current changes from 20 mA to 64 mA as the base current
increases one 0.1 mA step. Thus ΔIc = -20 mA - (-64mA) = 44 mA. β is thus 44 mA
/ 0.1 mA = 440.
As the 2N4403's collector current increases, β declines, as is typical for a
transistor. In this case, from 440 to 60 over the current range 40 mA to 180 mA.
|
 |
The plot and data table below is extracted from Fairchild's 2N4403 data
sheet and confirms our measurement shape generally, but with much less drop in β
than measured. There are several reasons for this divergence.
First, β has great sample-to-sample variation, with the data sheet showing
the minimum β at 150 mA being 100 and the maximum β of 300. Our measured value
of 140 at 140 mA fits nicely within this rather broad range.
Second, beta is a strong function of junction temperature. The datasheet
values in the table are for 25°C, room temperature but if the junction
temperature is 125°C, β is greater for lower currents but falls off more rapidly
for higher currents. This is exactly what the measurements show. And, indeed, it
is clear that at the current levels used in the test data, the junction
temperature is elevated well above room temperature.
| |
Datasheet (25°C) |
Measured |
|
Ic (mA) |
Raw Value |
Normalized |
Raw Value |
Normalized |
|
40 |
220 |
1.000 |
440 |
1.000 |
|
140 |
185 |
0.841 |
140 |
0.318 |
|
180 |
170 |
0.773 |
60 |
0.136 |
|
|
|
|
2N4403 hFE versus collector current. From Fairchild 2N4403 data sheet.
|
 |
I'll add a web page in the next week or two exploring transistor
parameters as measured by the curve tracer in more detail. And, I'll use NPN
devices. I used PNP parts because the 575's manual's examples are almost
exclusively PNP transistors and I was trying to duplicate the curves and
parameter settings whilst learning my way around the 575. (And why did the 575
show almost exclusively PNP transistors while today the majority of transistors
used are NPN? It turns out that in the early days of transistor development, it
was much easier to build functioning PNP transistors with better performance
than NPNs. That's no longer the case, but it certainly was in the late 1950's
when the 575 was developed.) |
|
19 May 2009
I've added Google's internal search function to the site.
It should help locate the information scattered quasi-randomly across the site.
The search box appears near the top of all pages.
|
|
18 May 2009
I've measured the Z10040A Norton Amplifier's noise figure
today and revised the Norton Amplifier's page
accordingly. I measured the NF as between 3.5 and 3.7 dB at 30 MHz.
|
|
17 May 2009
I've added measured IP3 data for three new amplifiers as
well as IP3 versus supply voltage for the Z10040A Norton amplifier at
Measuring IP3.
|
|
14 May 2009
I've added a new page describing how I measure 3rd order
intercept, IP3, for preamplifiers, Measuring IP3.
Data includes a Z10000-U buffer amplifier and a Mini-Circuits ZFL-500LN
amplifier as well as the Z10040A Norton amplifier.
|
|
11 May 2009
I've added a new page describing how I measure second
order intercept, IP2, for preamplifiers, Measuring
IP2.
I've also added pricing and a photograph for the Z10040A
Norton amplifier indoor enclosure, more information at
Z10040A Norton Amplifier.
The photo below shows four Z10040A Norton amplifiers I
assembled over the weekend.
|
 |
|
09 May 2009
I've fixed an error on my
Metal Oxide Varistor (MOV) page.
Thanks to Dave, G3TJP for noting the discrepancy between an entry in the data
table and the text. It's now corrected (the V18ZA40P's 80J rating in the table
is correct and I've modified the text to reflect 80J.)
|
|
08 May 2009
I've revised the
Z10040A Norton Amplifier manual
to correct a few small errors and to reflect one part change. It also contains a
revised mounting hole drawing, supplying one dimension omitted in earlier
versions.
The new manual also includes a copy of Dr. Norton's
patent.
|
|
07 May 2009
I've added metal oxide varistors (MOV) as power supply protective devices
to several projects under development. So far, the only one that has made it
to an available kit is the Z10040A Norton
Amplifier, but they will appear in future kits.
All these kits require DC power in the 13-15 volt range, and my thought in
adding the MOV is not so much to protect the kit electronics from an inadvertent
of an over-voltage but rather to limit the damage a gross over-voltage condition
generated externally might cause to other equipment powered from the same
source.
I've created a new web page Metal
Oxide Varistor (MOV) showing how the MOV's I'm using react to over-voltage.
|
|
05 May 2009
I've revised the
Z10040A Norton amplifier manual
to correct a couple of small errors and expand the circuit description. The most
important change is four 1000pF capacitors were incorrectly identified as 0U01
(0.01uF), a holdover from earlier versions I had built.
I've also added a lengthy section to the
Z10040A Norton amplifier page explaining what
the various amplifier parameters actually mean in practical terms.
For Clifton Laboratories customers in the UK and other
European countries, Dave, G3TJP, is willing to assemble kits for purchasers in
exchange for a contribution to a suitable charity. Dave may be reached at
dave@lanks.freeserve.co.uk
|
|
03 May 2009
At last, the Z10040A Norton Amplifier kit is ready to
ship. The details are at Z10040A Norton Amplifier.
I have several updates to the site that have not
been made as I've concentrated on the Z10040A and another couple of new kits.
I'll make these updates over the next few days.
|
|
03 May 2009
As normal for the first of the month, I've moved the April
2009 updates to the archive, viewable by clicking here
or via the link table at the top of this page. |
